i386/75387: [ata] feature request: support of Promise SATAII150 TX4 wanted

[Søren Schmidt]

The following reply was made to PR i386/75387; it has been noted by GNATS.

From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= <sos at FreeBSD.ORG>
To: Mikhail Teterin <mi+mx at aldan.algebra.com>
Cc: bug-followup at FreeBSD.ORG, sebastian.holmqvist at gmail.com
Subject: Re: i386/75387: [ata] feature request: support of Promise SATAII150 TX4 wanted
Date: Wed, 16 Nov 2005 23:53:54 +0100

 On 16/11/2005, at 23:26, Mikhail Teterin wrote:
 
 > =D1=81=D0=B5=D1=80=D0=B5=D0=B4=D0=B0 16 =D0=BB=D0=B8=D1=81=D1=82=D0=BE=D0=
 =BF=D0=B0=D0=B4 2005 17:12, S=C3=B8ren Schmidt =D0=92=D0=B8 =20
 > =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BB=D0=B8:
 >> Hmm, in 6.0 and current the SATA2 promise chips are supported,
 >> however 3Gb is untested as I have no such equipment here. However,
 >> your chipid is unknown to me but you coud try to add it like the
 >> pdc40718/719 and let me know if that works.
 >>
 >> I'll ask promise about that chip, can you verify the chip type by
 >> reading whats printed on its back please ?
 >
 > It is already deep inside my desktop system -- I can't reach there...
 > The "marketing name" is SATA300 TX2plus (one PATA and 2 SATA =20
 > channels),
 > although TX4 can't be very different (it has 4 SATA channels =20
 > instead of 2):
 > After applying the patch, it is identified as:
 
 Well, the marketing bla bla doesn't tell the story nor the patch, I =20
 need to be sure whats on the chip, or get it verified by promise from =20=
 
 the pci id.
 >
 > IMHO, the driver should be smarter in identifying devices, you have =20=
 
 > not seen
 > before as compatible with something, you have. Anything unknown, =20
 > with Promise
 > being the vendor and 'mass storate' being the class should be =20
 > examined as an
 > ATA controller, for example.
 
 That wont fly sadly, if you look closer you'll see that things change =20=
 
 from generation to generation of the Promise chips, so chances are =20
 very slim it will work.
 
 With controllers saying they are ATA of some sort it will be treated =20
 as a generic chip, and it should work even with DMA but most likely =20
 with reduced performance. Promise chips are *not* in this category...
 
 S=C3=B8ren Schmidt
 sos at FreeBSD.org